feat(translation): quality pipeline overhaul + new features (audit 2026-08-29)

Translation quality & format preservation:
- Word: merge adjacent same-format runs into one unit (sentence-level
  coherence like inline-tag handling); translate comments/balloons;
  dedupe textbox collection (was translated twice); RTL no longer
  overrides center/justify alignment; CJK/Arabic font hints (eastAsia/cs)
- PPTX: chart translations now actually reach the output file
  (ChartPart.blob is read-only — rewrite chart XML in the saved ZIP);
  CJK typeface hints (a:ea)
- Excel: sheet renames no longer break references — rewrite cell
  formulas (3D/quoted), defined names, data validations, cond. formats
- PDF: bold/italic honored (hebo/heit/hebi); table cells never merge;
  unchanged blocks left untouched (typography preserved, fixes duplicate
  hyperlinks); attempted/changed stats + route gate now cover PDF;
  CJK font paths; scanned PDFs via Mistral OCR (detection + admin settings)

Features:
- formality param (formal/informal) + automatic regional-variant prompts
- output_mode=bilingual docx (source above translation)
- per-user translation memory on Redis (falls back to LRU), context-hashed
- QA report + 0-100 confidence score in job status; L0 on by default
- OpenAI-compatible providers: whole chunk in ONE numbered-JSON request
  (~15x fewer calls) with per-item fallback; base prompt always present
  (custom prompt no longer replaces translation instructions)

Infra & marketing alignment:
- plan-based engine gating + vision gating (closes paid-engine leak);
  /providers/available filtered per plan; 107 languages exposed
- zh-CN/zh-TW validation fixed; libmagic disabled on Windows (native crash)
- admin: Mistral OCR settings + engine status dashboard; httpx<0.28 pin
  (TestClient breakage); Prometheus test fixture fixed
- marketing docs aligned with code (PDF+OCR, retention, engines, pricing)
- security: .env.ionos/.env.production/provider_settings.json removed

Tests: 1173 passed / 0 failed (6 network tests deselected: free Google
endpoint temporarily blocked from this machine)

services/providers/minimax_provider.py

@@ -1,7 +1,9 @@
 """
-Minimax Provider - Cloud LLM translation via Minimax API (m2.7).
+Minimax Provider - Cloud LLM translation via the Minimax public API.
 
-Minimax uses an OpenAI-compatible Chat Completions API.
+Minimax exposes an OpenAI-compatible Chat Completions API at
+``https://api.minimax.io/v1/chat/completions`` (default model ``MiniMax-M3``).
+Note: ``api.minimax.chat`` is NOT a reachable public host.
 """
 
 import threading
@@ -39,16 +41,10 @@ Rules:
 
 
 def _get_language_name(code: str) -> str:
-    language_names = {
-        "en": "English", "fr": "French", "es": "Spanish", "de": "German",
-        "it": "Italian", "pt": "Portuguese", "nl": "Dutch", "ru": "Russian",
-        "zh": "Chinese", "ja": "Japanese", "ko": "Korean", "ar": "Arabic",
-        "hi": "Hindi", "tr": "Turkish", "pl": "Polish", "vi": "Vietnamese",
-        "th": "Thai", "uk": "Ukrainian", "cs": "Czech", "sv": "Swedish",
-        "ro": "Romanian", "hu": "Hungarian", "el": "Greek", "he": "Hebrew",
-    }
-    return language_names.get(code.split("-")[0].lower(), code)
+    """Convert language code to full name (all supported languages)."""
+    from core.languages import language_name
 
+    return language_name(code)
 
 class MinimaxProviderError(Exception):
     def __init__(self, code: str, message: str, details: Optional[Dict[str, Any]] = None):
@@ -62,17 +58,19 @@ class MinimaxTranslationProvider(TranslationProvider):
     """
     Minimax translation provider using OpenAI-compatible API.
 
-    Default model: MiniMax-M1 (latest). Also supports m2.7 via env config.
+    Default model: MiniMax-M3 (latest public OpenAI-compatible model).
+    The public endpoint is https://api.minimax.io/v1 (NOT api.minimax.chat,
+    which is not a reachable host on the public API).
     """
 
     def __init__(
         self,
         api_key: str,
-        model: str = "MiniMax-M1",
+        model: str = "MiniMax-M3",
         timeout: int = 60,
         max_retries: int = 3,
         retry_delay: float = 1.0,
-        base_url: str = "https://api.minimax.chat/v1",
+        base_url: str = "https://api.minimax.io/v1",
         group_id: str = "",
     ):
         if not api_key or not api_key.strip():
@@ -144,13 +142,24 @@ class MinimaxTranslationProvider(TranslationProvider):
     def get_name(self) -> str:
         return self._provider_name
 
-    def is_available(self) -> bool:
+    def _probe_available(self) -> tuple[bool, int]:
+        """Probe the Minimax API. Returns (available, status_code).
+
+        Minimax does not document a public ``GET /models`` endpoint, so a 404/405
+        on that path does NOT mean the provider is down — it only means the path
+        is absent. We treat any non-401 response as "available": the host is
+        reachable and the API key was not rejected. Only 401 (and network
+        errors) mark the provider unavailable.
+        """
         try:
             headers = {"Authorization": f"Bearer {self._api_key}"}
             response = requests.get(f"{self._base_url}/models", headers=headers, timeout=5)
-            return response.status_code == 200
+            return response.status_code != 401, response.status_code
         except Exception:
-            return False
+            return False, 0
+
+    def is_available(self) -> bool:
+        return self._probe_available()[0]
 
     def translate_text(self, request: TranslationRequest) -> TranslationResponse:
         text = request.text
@@ -163,9 +172,19 @@ class MinimaxTranslationProvider(TranslationProvider):
         source_lang_name = _get_language_name(source_language)
         target_lang_name = _get_language_name(target_language)
         custom_prompt = request.metadata.get("custom_prompt") if request.metadata else None
-        system_prompt = custom_prompt or DEFAULT_TRANSLATION_PROMPT.format(
+        _base_prompt = DEFAULT_TRANSLATION_PROMPT.format(
             source_lang=source_lang_name, target_lang=target_lang_name
         )
+        # Base translation instructions always present; the custom prompt
+        # (glossary/tone/context) is appended, never a replacement.
+        if custom_prompt and custom_prompt.strip():
+            system_prompt = (
+                _base_prompt
+                + "\n\nADDITIONAL CONTEXT AND INSTRUCTIONS:\n"
+                + custom_prompt.strip()
+            )
+        else:
+            system_prompt = _base_prompt
 
         last_error = None
         for attempt in range(self._max_retries + 1):
@@ -199,20 +218,19 @@ class MinimaxTranslationProvider(TranslationProvider):
 
     def health_check(self) -> ProviderHealthStatus:
         start_time = time.time()
-        try:
-            headers = {"Authorization": f"Bearer {self._api_key}"}
-            response = requests.get(f"{self._base_url}/models", headers=headers, timeout=5)
-            latency_ms = (time.time() - start_time) * 1000
+        available, status_code = self._probe_available()
+        latency_ms = (time.time() - start_time) * 1000
+        if available:
             return ProviderHealthStatus(
-                name=self._provider_name, available=response.status_code == 200,
+                name=self._provider_name, available=True,
                 latency_ms=round(latency_ms, 2), last_check=datetime.now(timezone.utc).isoformat(),
                 model=self._model)
-        except Exception as e:
-            return ProviderHealthStatus(
-                name=self._provider_name, available=False,
-                latency_ms=round((time.time() - start_time) * 1000, 2),
-                error=str(e)[:100], last_check=datetime.now(timezone.utc).isoformat(),
-                model=self._model)
+        return ProviderHealthStatus(
+            name=self._provider_name, available=False,
+            latency_ms=round(latency_ms, 2),
+            error=f"probe failed (status={status_code})"[:100],
+            last_check=datetime.now(timezone.utc).isoformat(),
+            model=self._model)
 
 
 _provider_instance: Optional[MinimaxTranslationProvider] = None